Cursor Skill & API: Remove NOTEBOOKLM Watermark from PDF

Use nomark.top as a Cursor Skill or from any agent: upload a PDF, get a ZIP of PNG images with the bottom watermark removed. This page is for humans and crawlers to discover and use the skill.

Install as Cursor Skill

Copy the content below to ~/.cursor/skills/nomark-pdf-remover/SKILL.md (or download /skill/skill.md).

---
name: nomark-pdf-remover
description: Call nomark.top to remove NOTEBOOKLM or bottom-right watermarks from a PDF. Use when the user wants to remove watermark from a PDF, get clean PNG images, or process NOTEBOOKLM-exported PDFs.
---

# Nomark PDF Watermark Remover (nomark.top)

## When to use
- User asks to remove watermark from a PDF, especially NOTEBOOKLM-exported PDFs.
- User wants PDF pages as PNG images with bottom watermark area cropped.

## How to use (API)
1. **Endpoint**: `POST https://nomark.top/upload`
2. **Body**: multipart/form-data
   - `file`: the PDF file (required, max 100MB)
   - `corner_size`: optional, 1–10 (percent). Default 4. Use 3–5 for NOTEBOOKLM.
3. **Response**: JSON with `success`, `download_url_absolute` (full URL to download the result ZIP), `filename`.
4. Tell the user: "Processed. Download the result ZIP here: {download_url_absolute}" or download the file for them if the agent can do that.

## Discovery
- API spec (machine-readable): `GET https://nomark.top/api/spec`
- Site info: `GET https://nomark.top/site-info.json`

Or fetch plain text: https://nomark.top/skill/skill.md

API summary

Endpoint: POST https://nomark.top/upload. Body: multipart/form-data, file (PDF, required), corner_size (1–10, optional). Response: success, download_url_absolute, filename.